FeaturesPipeline
- class lightautoml.pipelines.features.base.FeaturesPipeline(**kwargs)[source]
Bases:
object
Abstract class.
Analyze train dataset and create composite transformer based on subset of features. Instance can be interpreted like Transformer (look for
LAMLTransformer
) with delayed initialization (based on dataset metadata) Main method, user should define in custom pipeline is.create_pipeline
. For example, look atLGBSimpleFeatures
. After FeaturePipeline instance is created, it is used like transformer with.fit_transform
and.transform
method.- property input_features
Names of input features of train data.
- property output_features
List of feature names that produces _pipeline.
- property used_features
List of feature names from original dataset that was used to produce output.
- create_pipeline(train)[source]
Analyse dataset and create composite transformer.
- Parameters:
train (
LAMLDataset
) – Dataset with train data.- Return type:
- Returns: # noqa DAR202
Composite transformer (pipeline).
- fit_transform(train)[source]
Create pipeline and then fit on train data and then transform.
- Parameters:
train (
LAMLDataset
) – Dataset with train data.- Return type:
- Returns:
Dataset with new features.
- transform(test)[source]
Apply created pipeline to new data.
- Parameters:
test (
LAMLDataset
) – Dataset with test data.- Return type:
- Returns:
Dataset with new features.